python之moviepy库的安装与使用_python中如何安装moviepy.editor

目的：因为需要保存一个大大的.mp4视频，以防过程中设备出现异常导致整个长长的视频无法正常保存，所以采用分段保存视频的方式，每500帧保存一段，然后再将视频合到一起．最近刚开始学习python,发现python真的很好用，所以这次就使用python中的moviepy库来完成视频的合并．

一．安装moviepy

1.　你首先尝试使用 pip install moviepy指令是否可以正常安装moviepy库（我在python2.7上和python3.7上都尝试了这中安装方式都安装不了，所以不得不采用下面这个方式）

２．采用source文件安装．参照　https://blog.csdn.net/ucsheep/article/details/81000982　下载这个库的source文件，然后按照目录下的　README.rst　的指示安装，

首先cd到你下载的目录文件下，顺序执行　

$ (sudo) pip install ez_setup
$　(sudo) python setup.py install

 

如果有错误提示

ERROR: moviepy 1.0.3 has requirement imageio<2.5,>=2.0, but you'll have imageio 2.8.0 which is incompatible.

那么就执行

pip install imageio

二．使用moviepy库合并多个视频

我的目录框架是这样的，

 

上面的每个文件目录下的文件是这样的．

 Bluetooth文件目录

 wifi文件目录

 首先是将目录下的所有视频都合并到一起

下面的代码实现视频的合并，文件的合并，多个文件夹下文件的归并

# -*-coding:utf-8-*-
# this python script is to concatenate a sequence of videos into one
 
 
# import cv2
 
from moviepy.editor import *
import os
import linecache
import shutil
 
#inputVideoPath = "/media/yunlei/Seagate/collection_device_data/20200603/大仟里L2主干道大圈-1591150453197/"
# outputVideoPath = "/media/yunlei/Seagate/collection_device_data/20200603/concatenated/大仟里L2主干道大圈-1591150453197/"
inputVideoPath = "/media/yunlei/Seagate/collection_device_data/20200603/大仟里L2主干道-1591155992285/"
 
outputVideoPath = "/media/yunlei/Seagate/collection_device_data/20200603/concatenated/大仟里L2主干道-1591155992285/"
 
 
def videoconcatenate(left_right_depth):
    print("this function is to implement concatenation")
    dirs = os.listdir(inputVideoPath)
    videoList = []
    videoCount = 0
    for videoDir in dirs:
        videoName = inputVideoPath + str(videoCount) + "/" + left_right_depth + ".mp4"
        if os.path.exists(videoName):
            videoElement = VideoFileClip(videoName)
            videoList.append(videoElement)
        videoCount = videoCount + 1
    concatenateProcessLeft = concatenate_videoclips(videoList)
    concatenateProcessLeft.to_videofile(outputVideoPath + "/" + left_right_depth + ".mp4", fps = 20,  remove_temp = False)
 
def combinefiles(fileName):
    print("start to comebine files")
    #This list is used to store all file data
    filePathList = []
    fileDataList = []
 
    fileCount = 0
 
    filePathes = os.listdir(inputVideoPath)
    for filePath in filePathes:
        fileType = inputVideoPath + str(fileCount) + "/" + fileName + ".txt"
        if os.path.exists(fileType):
            filePathList.append(fileType)
            # print(fileType)
        fileCount = fileCount + 1
 
    totalline = 0
    for fileElement in filePathList:
        lineNumber = 1
        fileLength = len(open(fileElement, encoding='utf-8').readlines())
        totalline = totalline + fileLength
        #print(fileLength)
 
        while lineNumber <= fileLength:
            line = linecache.getline(fileElement, lineNumber)
 
            #print(line)
            line = line.strip()
            fileDataList.append(line)
            lineNumber = lineNumber + 1
    print(totalline)
 
    fileAll = open(outputVideoPath + "/" + fileName + ".txt", 'w+', encoding='utf-8')
    for i, p in enumerate(fileDataList):
        print(i,p)
        fileAll.write(p+'\n')
    fileAll.close()
 
def combineFolders(folderName):
    folderList = []
    folderCount = 0
    outputFolderPath = outputVideoPath + "/" + folderName + "/"
    folderPathes = os.listdir(inputVideoPath)
    print(folderPathes)
    for folderPath in  folderPathes:
        folerType = inputVideoPath + "/" + str(folderCount) + "/" + folderName
        print("folderType")
        print(folerType)
        print("start to copy file")
 
        if os.path.exists(folerType):
            filesInFolder  = os.listdir(folerType)
            print("filesInFolder")
            print(filesInFolder)
            for fileInFolder in  filesInFolder:
                totalPath = folerType + "/" + fileInFolder
                print("print totalPath")
                print(totalPath)
                if not os.path.exists(outputFolderPath):
                    os.mkdir(outputFolderPath)
                outputFileName = outputFolderPath + "/" + fileInFolder
                shutil.copyfile(totalPath, outputFileName)
        folderCount = folderCount + 1
 
 
 
 
#define the main function,from this function your users functions are called
def main():
    # combine Bluetooth folder
    combineFolders("Bluetooth")
    # combine wifi folder
    combineFolders("wifi")
    # concatenate video left
    videoconcatenate("left")
    # # # concatenate video right
    videoconcatenate("right")
    # # #concatenate video depth
    videoconcatenate("depth")
    # combinefiles("video_time")
    combinefiles("Camera_time")
    # combinefiles("Bluetooth_times")
    combinefiles("wifi_times")
#the entrance of this projrct
if __name__ == "__main__":
    main()
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 

因为是刚学习python所以很多时候并不知哪个用法更合适，所以那就尝试一下，比如下面这两个遍历路径下的文件的方式，

for videoDir in dirs:

这种，会将dirs路径下的所有文件都获取到，如果比如说我这里的路径下就包括了文件加和文件，而我希望对文件夹做处理，所以我就要先将文件夹挑拣出来．下面就是我只检索那些是文件夹，并且文件夹上有.mp4格式视频的文件目录我才把他们count in.

    videoLeft = inputVideoPath + str(videoCount) + "/" + "left.mp4"
    videoRight = inputVideoPath + str(videoCount) + "/" + "right.mp4"
    videoDepth = inputVideoPath + str(videoCount) + "/" + "depth.mp4"
    if os.path.exists(videoLeft):

第二种，这种os.walk(path)的方式可以返回root就是根目录path，dirs就是root目录下所有的文件夹，以及文件夹下的文件夹，files就是root path下所有的文件．所以你需要根据你的需求来选择使用哪种遍历方式．

for root, dirs, files in os.walk(inputVideoPath):
	for name in files:
		print(os.path.join(root, name))
	for name in dirs:
		print(os.path.join(root, name))
print(len(videoLeftAll))